The carbocationic polymerization of isobutylene (IB) is the subject of great scientific and industrial interest. The unique properties of polyisobutylene (PIB), a chemically stable fully saturated polymer make it a desirable material with applications ranging from medical devices to ashless (metal-free) dispersants/detergents suitable for use as motor oil and fuel additives. These ashless dispersants/detergents can be characterized as oil soluble surfactants with oligoamine end-groups derived from low molecular weight (number average molecular weight ( Mn) of from about 500 to about 5000) PIB or polybutenes (copolymers of IB with C4 olefins) having olefinic end groups.
Two major industrial methods have been developed to produce low molecular weight IB homo or copolymers with olefinic end groups. The “conventional” method uses a C4 mixture and an aluminum halide based catalyst system and produces polybutenes with high tri-substituted olefinic contents. Due to the low reactivity of the tri-substituted olefinic end groups, polybutenes need to be chlorinated to react with maleic anhydride to give polybutenylsuccinic anhydride, which is subsequently reacted with oligoalkylenimines to yield polybutenylsuccinimide-type ashless dispersant/detergent. The other method employs a pure IB feed stream and a BF3 complex-based catalyst with either alcohols, or ethers in a polymerization reaction run at low temperature, which yields highly reactive PIB (HR PIB) with high exo-olefinic end-group contents. In contrast to the tri-substituted olefins of conventional polybutenes, PIB exo-olefins readily react with maleic anhydride in a thermal “ene” reaction to produce PIB succinic anhydride and subsequently polyisobutenylsuccinimide ashless dispersants. Because the final product does not contain chlorine, HR PIB is more desirable than conventional polybutenes. However, BF3 is difficult to handle and the polymer may contain fluorine. Further, as noted above, this method requires a pure IB feed steam and low temperature (e.g., −30° C.) and therefore results in a more expensive product.
The above-described commercial process for producing HR PIB has been reported by U.S. Pat. No. 5,408,018 (and DE-A 2702604) to Rath. A range of process enhancements were subsequently reported in. U.S. Pat. Nos. 6,407,186, 6,753,389, and 7,217,773 to Rath et al. and U.S. Pat. Nos. 6,846,903, 6,939,943 and 7,038,008 to Wettling et al. A modified process using a different temperature regime and a low residence time was also previously described (e.g., U.S. Pat. Nos. 6,562,913 and 6,683,138 to Baxter et al.). All of these disclosures describe polymerizations carried out with BF3 catalyst and an alcohol or ether co-catalyst. Such catalytic processes can leave residual fluorine in the polymer especially when utilized with the commonly available mixed C4 Raffinate I stream. The presence of even small amounts of fluorine cause problems in downstream functionalization reactors due to the release of HF and therefore require expensive fluorine removal post-treatment.
Many attempts have therefore been made to find other methods for producing HR PIB. For instance PIB with nearly quantitative exo-olefin end groups has been obtained by reacting tert-chloride-terminated PIB (PIB-Cl) with strong bases such as potassium tert-butoxide and alkali ethoxides in refluxing tetrahydrofuran (THF) for 20-24 h, (Kennedy, J. P.; Chang, V. S. C.; Smith, R. A.; Iván, B. Polym. Bull. 1979, 1, 575); quenching living PIB with methallyltrimethylsilane, (Nielsen, L. V.; Nielson, R. R.; Gao, B.; Kops, J.; Iván, B. Polymer 1997, 38, 2528); quenching living PIB with a hindered base (e.g., 2,5-dimethylpyrrole or 1,2,2,6,6-pentamethylpiperidine), (Simison, K. L.; Stokes, C. D.; Harrison, J. J.; Storey, R. F. Macromolecules 2006, 39, 2481); quenching living PIB with an alkoxysilane or an ether compound (Storey, R. F.; Kemp, L. L. U.S. Patent Application Publication, 2009/0318624 A1, Dec. 24, 2009); and reacting living PIB with a mono-sulfide followed by decomposition of the resulting sulfonium salt with a base (Morgan. D. L.; Stokes, C. D.; Meierhoefer, M. A.; Storey, R. F. Macromolecules 2009, 42, 2344). However, all of the above methods are expensive as they involve living cationic polymerization at low temperature in a moderately polar solvent, and employ expensive reactants.
A broad disclosure of halogen-free metal catalysts based on oxides of Groups V and VI of the Periodic Table of Elements was described in U.S. Pat. No. 6,441,110 to Sigwart et al., but these catalysts were heterogeneous and gave poor monomer conversions, with only modest amounts of exo-olefins. Another catalyst system, based on metals from the 3rd to the 12th periods of the periodic system of elements with nitrile ligands and weakly coordinating anions was described in U.S. Pat. No. 7,291,758 to Bohnepoll et al. These catalysts were used only in a polar dichloromethane solution; not in an apolar, all-hydrocarbon media.
More recently it has been reported that AlCl3-OBu2 complexes in conjunction with a range of initiators or adventitious water initiate the polymerization of IB and in polar solvent (CH2Cl2/hexane 80/20 v/v) yield PIB with high exo-olefinic end groups up to 95% in a range of temperatures (−60 to −20° C.) (Vasilenko, I. V.; Frolov, A. N.; Kostjuk, S. V. Macromolecules 2010, 43(13), 5503-5507). Independently, similar results were reported with adventitious water as initiator in conjunction with AlCl3 or FeCl3 dialkyl ether complexes in CH2Cl2 at temperatures ranging from −20 to 20° C. (Lui, Q.; Wu Y.; Zhang, Y.; Yan. P. F.; Xu, R. W. Polymers 2010, 51, 5960-5969). However, due to the need for the polar solvent CH2Cl2 the commercial potential of this method is questionable. AlCl3-OBu2 has been reported to produce PIB with terminal vinylidene bonds in the absence of solvent and without added initiator, or with water as an added initiator (USPG 2011/0201772A1 of König et al.). However, none of the conventional cationic initiators such as alkyl halides, ethers, esters, alcohols and Brønsted acids were found to initiate directly the polymerization in apolar media with AlCl3. Therefore there remains a need for a robust and economic method for the preparation of highly reactive PIB or polybutene in a non-polar hydrocarbon media.
It was previously found that conventional cationic initiators used in conjunction with certain combinations of Lewis acid/Lewis base complexes initiate the polymerization of IB in hydrocarbon solvents at temperatures of −30° C. to +50° C. and provide high yields of highly reactive PIB. More specifically, it was recognized that polymerization of IB in hydrocarbon solvents at temperatures of −30° C. to +50° C. could be initiated, and provide high yields of highly reactive PIB with a catalyst-initiator system comprising a Lewis acid catalyst (MR″Yn) complexed to a Lewis base (B) in an apolar medium.
The affinity of the Lewis acid for the Lewis base was such that the Lewis acid can interact at least partially with a conventional cationic initiator (RX), enabling the formation of R+, which initiated cationic polymerization of isobutylene. The initially complexed Lewis base was capable of effecting rapid deprotonation of the growing carbocation chain to form an exo-olefin prior to isomerization of the cationic chain or diffusion of the base into free solution. If the Lewis base did not complex to the Lewis acid, at least initially, deprotonation of the growing carbocation was not sufficiently rapid relative to isomerization to give the desired high yields of exo-olefin. If the interaction of the Lewis acid and the Lewis base was so strong that it prevented interaction with a conventional initiator, polymerization was be inefficient or did not occur at all. The Lewis acid or Lewis acid-Lewis base complex had to further be capable of interacting with a conventional initiator either in the presence or absence of adventitious water. Catalysts with which monomer conversion is substantially dependent on adventitious water, even in the presence conventional initiators, as evidenced by complete loss of conversion in the presence of a proton trap (e.g., 2,6-ditert-butylpyridine or “DTBP”), were not suitable. In the above systems, suitable Lewis bases were oxygen- and/or sulfur-containing nucleophiles, such as ethers, alcohols, ketones, aldehydes and esters, and sulfur-containing nucleophiles such as thioethers and thioketones.
Although the above polymerization system was found to provide excellent yields of highly reactive PIB, further study found that the yield could be further increased by proper selection the Lewis base and the control of the ratio of Lewis acid to Lewis base used to form the complex.